SPDNA: A User Recommendation Model for Academic Social Network

In academic social networks, each user's personal attributes and behaviour preferences form a unique social gene. The recommendation of friends in academic social networks is to use scholars' social genes to recommend scholars who have similar social genes with that scholar. In order to consider the social network structure of users and the attributes of users themselves, SPDNA (Scholar Profile DNA) model is proposed in this paper. The SPDNA model is the user attributes, user preference factor will be extracted according to the DNA model, while the user is formed in the academic social network's influence as the user influence factor, SPDNA factor is used to measure the size of the carrier. The constructed SPDNA model can reflect the personal attributes and behavior preferences of users in academic social networks, and also consider the influence of users in social networks. In the recommended for the user, according to each user to get the SPDNA string matching factor, then the various factors of the SPDNA string similarity calculation summation, then the similarity value obtained in TopK sequencing, end user recommendation set by user influence factor threshold filtering. In addition, through experiments and comparisons, the SPDNA model proposed in this paper also achieves good results in the recommendation of friends, and provides a new solution for the recommendation of friends in academic social networks.